Resumen de Towards a more flexible, sustainable, efficient and reliable induction cooking: a power semiconductor device perspective
This thesis is focused on addressing a more flexible, sustainable, efficient and reliable induction cooking approach from a power semiconductor device perspective. In this framework, this PhD Thesis has identified the following activities to cover such demands:
In view of the growing interest for an effective power multiplexing in Induction Heating (IH) applications, improved and efficient Solid State Relays (SSRs) as an alternative to the electromechanical relays (EMRs) are deeply investigated. In this context, emerging Gallium Nitride (GaN) High‐Electron‐Mobility Transistors (GaN HEMTs) and Silicon Carbide (SiC) based devices are identified as potential candidates for the mentioned application, featuring several improved characteristics over EMRs. On the contrary, other solutions, which seemed to be very promising, resulted to suffer from anomalous behaviors; i.e. SJ MOSFETs are thoroughly analysed by electro‐thermal physical simulations at the device level.
Additionally, the Short Circuit (SC) capability of power semiconductor devices employed or with potential to be used in IH appliances is also analysed. On the one hand, conventional IGBTs SC behavior is evaluated under different test conditions so that to obtain the trade‐off between ruggedness and low power losses. Moreover, ruggedness and reliability of several normally‐off 600‐650 V GaN HEMTs are deeply investigated by experimentation and physics‐based simulation.
Finally, power losses calculation at die‐level is performed for resonant power converters by means of using Infrared Thermography (IRT). This method assists to determine, at the die‐level, the power losses and current distribution in IGBTs used in resonant soft‐switching power converters when functioning within or outside the Zero Voltage Switching (ZVS) condition. As a result, relevant information is obtained related to decreasing the power losses during commutation in the final application, and a thermal model is extracted for simulation purposes.
